Overview of Speech Translation at ATR

A speech translation system will transform a spoken dialogue from the speaker's language to the listener’s automatically and simultaneously. It will undoubtedly be used to overcome language barriers and facilitate communication among the peoples of the world. Creation of such a system will first require developing the various constituent technologies: speech recognition, machine translation, and speech synthesis. These individual subsystems will then be integrated to form a speech translation system, moreover an automatic telephone interpretation system. In Japan, the general view toward the research and development for an automatic telephone interpretation system was reported in 1986. It reported the necessity of long-term research and development in component technologies and software architecture. ATR Interpreting Telephony Research Laboratories was established in 1986 to initiate basic research for automatic telephone interpretation. Currently, ATR Interpreting Telephony Research Laboratories are engaged in research in the areas of continuous large-vocabulary Japanese speech recognition, integrated processing of speech and language, machine translation of spoken language from Japanese to English, and high-quality speech synthesis. Experimental spoken language translation system (SL-TRANS) was developed at ATR.